CAJ | 학술논문

基于多主体的建模仿真方法，运用particle swarm Optimization（PSO）群体智能算法模拟信息交互条件下外部投资者估价变化的学习机制和演化规律，在机制设计的基础上，建立了实现风险投资退出的股权拍卖模型．在Swarm平台上对股权拍卖模型的仿真分析表明，所设计的股权拍卖机制能够显著地提高风险投资家的收益，并能帮助风险投资家预测外部投资者的估价和拍卖参与度的变化．对股权拍卖模型的参数仿真发现，风险投资家可以通过弓j入更多的外部投资者参与股权拍卖来进一步提高自己的收益；即便外部投资者过度强化单一学习能力，最终也可以得到相对理想的股权拍卖结果．本文的研穿可以为风险投资家的策略选择提供参考依据。
기우다주체적건모방진방법，운용particle swarm Optimization（PSO）군체지능산법모의신식교호조건하외부투자자고개변화적학습궤제화연화규률，재궤제설계적기출상，건립료실현풍험투자퇴출적고권박매모형．재Swarm평태상대고권박매모형적방진분석표명，소설계적고권박매궤제능구현저지제고풍험투자가적수익，병능방조풍험투자가예측외부투자자적고개화박매삼여도적변화．대고권박매모형적삼수방진발현，풍험투자가가이통과궁j입경다적외부투자자삼여고권박매래진일보제고자기적수익；즉편외부투자자과도강화단일학습능력，최종야가이득도상대이상적고권박매결과．본문적연천가이위풍험투자가적책략선택제공삼고의거。
Based on multi-agent modeling and simulation method, this paper uses particle swarm opti- mization （PSO） swarm intelligence algorithm to simulate the learning mechanism and evolution rule of outside investors＇ valuations under the condition of information interaction, and establishes an equity auc- tion model of venture capital exit on the basis of mechanism design. Simulation analysis of equity auction model on Swarm platform shows that the above-mentioned equity auction mechanism can significantly improve the revenue of venture capitalist, and help venture capitalist to predict the change of outside investors＇ valuations and auction participation. Simulations of equity auction model＇s parameters find that venture capitalist can further improve their income by introducing more outside investors to participate in the auction; even outside investors overemphasize the importance of one single learning ability, the relatively satisfactory equity auction results also can be achieved eventually. The study of this article can provide policy references for venture capitalist.